Statistical Control Charts: What they are and what they are for

The control charts They are an aid used for the verification of the various processes that may occur in a theoretical or experimental work scheme, which is represented by diagrams. In them, an effective monitoring of the deviation or variation of the line of events studied can be carried out.

These types of representations are commonly used in many study disciplines, being a medium or statistical method that helps users prevent production problems, future malfunctions and many other events, generating a confidence interval in which you can work without any problem.

What are control charts for?

Among the most common areas in which this type of graphics are used are:

• Diagnosis of the behavior of events in a period of time determined or established by the user.
• Control of the processes carried out under any activity of interest that allows it to be charted.
• Determination of the influencing factors in the variation of the graphs made, which lead to a change in the trend line drawn.
• Appreciation and diagnosis of possible events and trend lines that may arise along a given time line.
• Improve the capacity of the machines and production processes of the companies.
• Strengthen regulations and production processes that lead to quality and efficiency.
• Apply means of measurement that are perfectly adapted, teaching and training staff if it were the case, so that circumstantial problems or that may have permanence are avoided.

Control limits

They are a series of controls imposed by the users who study a specific case, who are in charge of keeping a full surveillance of the possible changes that can be carried out, therefore, when these limits are exceeded, it could be concluded that the case presented could have been control.

The control limits are horizontal lines establishedin the graphs and both are generated in the part higher (largest accepted value), such as lower (Smallest accepted value), based on the random variation that may arise within the known parameters, having in turn a linecentral which represents the midpoint (most acceptable value) between the aforementioned limits and where the trend of expected events is assumed.

Assignable causes

Are the specific causes of process instability determined, which leads to the variation of the graphs which can represent real events in themselves, being able to being is established as an assignable cause that can be corrected in time to avoid the famous spikes in graphics.

Random causes

These are events that mostly cannot be controlled, and they are events that eventually occur due to natural causes, being also small variations generated by the processes, which have nothing to do with human error. However, they can function as a point to generate a logical line of prediction of future events.

Variable control charts

It is the mathematical model that acquires an infinity of interval values, measured in a given interval of time, which can be called continuous variable data. These values ​​are those that They have a unit of measurement such as time, pressure, humidity, diameters, lengths and many more.

These types of graphs present a certain sensitivity that others cannot offer and are difficult to perceive, that is, they tend to have a constant trend so Any error or alteration can be perceived quickly, and the root of the problem can be captured for quick assistance and recovery of the effectiveness of the object of study.

You can also get graphs of:

• Averages and ranges.
• Averages and standard deviations.
• Medians and ranges.
• Individual readings and ranges.

Attribute control charts

Its about group of data that can be enumerated or counted, and they are simply determinable to be involved in a specific field of study such as defects in the production of some item in mass, number of defective items, number of mistakes made by hand human.

This type of graphs have the advantage of quickly synthesizing all the information collected by the user who studies the case, having various applications, since allows to classify whether said process is acceptable or not, which is why it is commonly implemented in mass production systems, to achieve an efficient process with the minimization of mistakes.

You can also get graphs of:

• Percentage of defective units or processes.
• Percentage of defects by work area.
• Number of units of possible defective areas.

How to make a control chart?

There are various softwares that are responsible for the realization of this type of graphics, however, every program is useless without the human hand that knows how to operate it with full knowledge, so that it is important to know a series of parameters that will generate a logical line of ideas for the effective and correct realization of this type of instruments of great help, having So:

• Determination or knowledge of the subject to be studied, in order to obtain the best results expected, taking into account whether it is a unit of measurement such as mass, volume, length or others. Collecting the necessary data (at least 20) to make the graph.
• Choose the type of control chart, taking into account the data collection carried out and the subject being discussed.
• Estimate the times of each process, taking into account the data collection and the step size that will be estimated for the graph, in order to establish the fewest possible errors.
• Determine the central trend line and the upper and lower limits as a security system based on the common behavior of the process knowing the possible assignable and random causes that can happen and are modifiers of the graphic.
• Represent the data obtained.
• Process of analysis of results and interpretation of the lines obtained.